We address the speaker independent automatic recognition of spontaneous speech in highly variable noise by applying semisupervised sparse non-negative matrix factorization (NMF) for speech enhancement coupled with our recently proposed frontend utilizing bottleneck (BN) features generated by a bidirectional Long Short-Term Memory (BLSTM) recurrent neural network. In our evaluation, we unite the...

This paper describes the considerations behind the design of the programming language Edison (Brinch Hansen 1981). Edison is deeply influenced by the advances made by Pascal (Wirth 1971) and its successors Concurrent Pascal (Brinch Hansen 1975) and Modula (Wirth 1977). In my attempt to learn from these previous efforts and improve on them I had to focus on both the virtues and defects of these ...

The most common types of nucleotide sequence data analyses and handling can be done more conveniently and inexpensively on microcomputers than on large time-sharing systems. We present a package of computer programs for the analysis of DNA and RNA sequence data which overcomes many of the limitations imposed by microcomputers, while offering most of the features of programs commonly available o...